A formulator's guide to sweet proteins

A sweet protein works at milligram doses, behaves like a protein under heat and pH, and replaces sweetness only, so you build a system around it rather than dropping it in one for one. This guide is for R&D teams meeting UnSugar Sweet Protein for the first time. It covers how potent it is, how it tastes, how it holds up in process, what it does not do, and how to get a clean read from a first bench trial.

What is a sweet protein?

UnSugar Sweet Protein is a sweet protein, the kind of sweetness found in fruit. It is a small protein and its folded shape lets it switch on the sweet-taste receptor, the part of the tongue that senses sweetness. Because it is a protein, your body digests it the way it handles any dietary protein, and at the milligram levels used for sweetness it contributes no sugars and negligible calories. That describes the molecule, not a health benefit.

How potent is it, and how do I dose it?

The sweet protein is about 500 to 2,000 times sweeter than sugar by weight. So you dose milligrams where you used to weigh grams. The use rate is roughly one part sweet protein for every 1,000 parts of the sugar it replaces.

Here is the arithmetic made easy. Every sample bottle holds enough sweet protein to replace up to about 5 kilograms of sugar. You dose it in tiny amounts, so a little goes a long way.

Say a flavored water should taste as sweet as 40 grams of sugar per liter. You dose it in tiny amounts, so add a little, taste, and adjust. Small weighing errors turn into large sweetness swings, so dose from a solution rather than trying to weigh dry powder by hand at the bench.

How does it taste?

The sweet protein carries a clean, rounded sweetness that sits far closer to sugar than stevia or monk fruit. It also rounds out the bitterness and off-notes of stevia and monk fruit, which is one of its most useful jobs in a blend. It carries a gentle, clean sweetness on the finish, not the bitter or licorice tail of those two. Be honest with your sensory panel: it is not identical to sugar, so set that expectation up front and your panel stays calibrated.

How does it behave under heat and pH?

Water protects the sweet protein. High-moisture products hold about 80 to 85°C at the core whatever the oven is set to, so moist cakes, pasteurized drinks, sauces, and jams are fine. Long periods of high heat in low-moisture products, above about 90 to 100°C for a long time, can relax the protein so it tastes less sweet and the flavor shifts. That is a taste effect only, not a safety or quality concern. As a rule of thumb, it holds its sweetness up to about 98°C. The cases to watch are biscuits, crackers, extruded snacks, and high-temperature spray drying; for these we are developing a format for long, dry, high-temperature processes, so tell us your temperature, time and moisture and we will advise.

What does it not replace?

A sweet protein replaces sweetness. Sugar also does other work, and none of it comes along:

• Bulk and volume, the mass in candy, baked goods, and tabletop products.
• Browning, the browning that gives baked goods their crust and color.
• Texture and mouthfeel, body, chew, and freezing behavior in frozen desserts.
• Water activity, which affects shelf life.
• Yeast substrate, since in bread the yeast eats sugar.

So you build a system: the sweet protein for clean sweetness, plus allulose, fibers, or sugar alcohols to rebuild the mass and texture. Beverages, dairy, sports nutrition, and functional powders are the easiest wins because they lean on sweetness and masking. Confectionery, chocolate, and bakery are harder because bulk is central to what they are.

How do I run a first bench trial?

1. Pick a forgiving base. A drink or a simple flavored water removes the bulk problem and lets you read sweetness directly.
2. Dose in tiny amounts, not from dry powder. This is the biggest driver of repeatable results at the bench.
3. Bracket your target. Build three to five samples that step across the sweetness you are aiming for.
4. Match against a real reference, your full-sugar or current product, set beside the samples.
5. Rest and taste. Let taste trials rest 15 minutes before tasting, and have tasters note onset and finish, not just peak sweetness.
6. Then layer in the system, adding bulk and texture components and re-tasting.

Change one variable at a time. It is slower on paper and much faster in practice.

Start with a blend at about 65% of the sweetness from the sweet protein and 35% from your existing sweetener, and remember that final performance should be confirmed by bench testing in your own recipe.

When you are ready to test in your own matrix, Request Samples with the sample bottle and sample powder, or request the technical data sheet.